ewsr1 fusion proteins mediate pax7 expression in ewing...

9
EWSR1 fusion proteins mediate PAX7 expression in Ewing sarcoma Gregory W Charville 1 , Wei-Lien Wang 2 , Davis R Ingram 2 , Angshumoy Roy 3 , Dafydd Thomas 4 , Rajiv M Patel 4 , Jason L Hornick 5 , Matt van de Rijn 1 and Alexander J Lazar 2 1 Department of Pathology, Stanford University School of Medicine, Stanford, CA, USA; 2 Departments of Pathology & Translational Molecular Pathology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA; 3 Departments of Pathology & Immunology and Pediatrics, Texas Childrens Hospital, Baylor College of Medicine, Houston, TX, USA; 4 Department of Pathology, University of Michigan School of Medicine, Ann Arbor, MI, USA and 5 Department of Pathology, Brigham and Womens Hospital, Boston, MA, USA PAX7 is a paired-box transcription factor that is required for the developmental specification of adult skeletal muscle progenitors in mice. We previously demonstrated PAX7 expression as a marker of skeletal muscle differentiation in rhabdomyosarcoma. Here, using analyses of published whole-genome gene expression microarray data, we identify PAX7 as a gene with significantly increased expression in Ewing sarcoma in comparison to CIC-DUX4 round cell sarcoma. Analysis of PAX7 in a large cohort of 103 Ewing sarcoma cases by immunohistochemistry revealed expression in 99.0% of cases (102/103). PAX7 expression was noted in cases demonstrating three distinct Ewing sarcoma EWSR1 translocations involving FLI1, ERG, and NFATc2. No PAX7 expression was observed in any of 27 cases of CIC-DUX4 sarcoma by immunohistochemistry (0%; 0/27). Exploring the mechanism of PAX7 expression in Ewing sarcoma using curated RNA- and ChIP-sequencing data, we demonstrate that the EWSR1 fusion protein is required for PAX7 expression in Ewing sarcoma and identify a candidate EWSR1-FLI1-bound PAX7 enhancer that coincides with both a consensus GGAA repeat-containing binding site and a peak of regulatory H3K27 acetylation. Taken together, our findings provide mechanistic support for the utility of PAX7 immunohistochemistry in the diagnosis of Ewing sarcoma, while linking this sarcoma of uncertain histogenesis to a key transcriptional regulator of mammalian muscle progenitor cells. Modern Pathology advance online publication, 23 June 2017; doi:10.1038/modpathol.2017.49 Ewing sarcoma is a malignant small, round, blue-cell neoplasm, which most often presents in children and young adults. Ewing sarcoma is often described as morphologically ambiguous, consisting of cells with finely dispersed chromatin, inconspicuous nucleoli, and scant amounts of clear to amphophilic cyto- plasm, giving the appearance of a highly cellular, bluetumor on hematoxylin/eosin-stained histology. Owing to this non-specific cytomorphology, the diagnosis of Ewing sarcoma was historically one of exclusion, requiring a thorough evaluation for morphologic and molecular markers of other small, round, blue-cell tumors. 1 The classification of Ewing sarcoma was revolutionized with the identification of a unique t(11;22) translocation, which is found in approximately 8590% of Ewing sarcoma cases, including those arising in both bone and soft tissue. 24 This translocation produces an oncogenic EWSR1-FLI1 fusion, combining the 5transcriptional activation domain of EWSR1 with the 3ETS DNA- binding domain of FLI1 to yield a potent oncogene. 5 Several variant translocations have been implicated in the minority of cases not exhibiting the typical EWSR1-FLI1 fusion. These variants all pair the 5portion of EWSR1 with the 3DNA-binding domain of a transcription factor (Figure 1). Of these rarer variants, the majority involve other ETS family transcription factors, such as ERG (the most common minor variant), 6 ETV1, 7 E1AF (ETV4), 8 or FEV. 9 More recently, even rarer fusion variants involving transcription factors without an ETS domain, such as NFATC2, have been discovered. 10 Correspondence: Dr GW Charville, MD, PhD, Department of Pathology, Stanford University School of Medicine, 300 Pasteur Drive, Lane 235, Stanford, CA 94305-5324, USA. E-mail: [email protected] Received 4 March 2017; revised 4 May 2017; accepted 21 May 2017; published online 23 June 2017 Modern Pathology (2017), 1 9 © 2017 USCAP, Inc All rights reserved 0893-3952/17 $32.00 1 www.modernpathology.org

Upload: vantu

Post on 20-Sep-2018

223 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: EWSR1 fusion proteins mediate PAX7 expression in Ewing sarcomamed.stanford.edu/labs/vanderijn-west/Publications/modpathol201749a… · EWSR1 fusion proteins mediate PAX7 expression

EWSR1 fusion proteins mediate PAX7expression in Ewing sarcomaGregory W Charville1, Wei-Lien Wang2, Davis R Ingram2, Angshumoy Roy3,Dafydd Thomas4, Rajiv M Patel4, Jason L Hornick5, Matt van de Rijn1 andAlexander J Lazar2

1Department of Pathology, Stanford University School of Medicine, Stanford, CA, USA; 2Departments ofPathology & Translational Molecular Pathology, The University of Texas MD Anderson Cancer Center,Houston, TX, USA; 3Departments of Pathology & Immunology and Pediatrics, Texas Children’s Hospital,Baylor College of Medicine, Houston, TX, USA; 4Department of Pathology, University of Michigan School ofMedicine, Ann Arbor, MI, USA and 5Department of Pathology, Brigham and Women’s Hospital, Boston, MA,USA

PAX7 is a paired-box transcription factor that is required for the developmental specification of adult skeletalmuscle progenitors in mice. We previously demonstrated PAX7 expression as a marker of skeletal muscledifferentiation in rhabdomyosarcoma. Here, using analyses of published whole-genome gene expressionmicroarray data, we identify PAX7 as a gene with significantly increased expression in Ewing sarcoma incomparison to CIC-DUX4 round cell sarcoma. Analysis of PAX7 in a large cohort of 103 Ewing sarcoma cases byimmunohistochemistry revealed expression in 99.0% of cases (102/103). PAX7 expression was noted in casesdemonstrating three distinct Ewing sarcoma EWSR1 translocations involving FLI1, ERG, and NFATc2. No PAX7expression was observed in any of 27 cases of CIC-DUX4 sarcoma by immunohistochemistry (0%; 0/27).Exploring the mechanism of PAX7 expression in Ewing sarcoma using curated RNA- and ChIP-sequencing data,we demonstrate that the EWSR1 fusion protein is required for PAX7 expression in Ewing sarcoma and identify acandidate EWSR1-FLI1-bound PAX7 enhancer that coincides with both a consensus GGAA repeat-containingbinding site and a peak of regulatory H3K27 acetylation. Taken together, our findings provide mechanisticsupport for the utility of PAX7 immunohistochemistry in the diagnosis of Ewing sarcoma, while linking thissarcoma of uncertain histogenesis to a key transcriptional regulator of mammalian muscle progenitor cells.Modern Pathology advance online publication, 23 June 2017; doi:10.1038/modpathol.2017.49

Ewing sarcoma is a malignant small, round, blue-cellneoplasm, which most often presents in children andyoung adults. Ewing sarcoma is often described asmorphologically ambiguous, consisting of cells withfinely dispersed chromatin, inconspicuous nucleoli,and scant amounts of clear to amphophilic cyto-plasm, giving the appearance of a highly cellular,‘blue’ tumor on hematoxylin/eosin-stained histology.Owing to this non-specific cytomorphology, thediagnosis of Ewing sarcoma was historically one ofexclusion, requiring a thorough evaluation formorphologic and molecular markers of other small,round, blue-cell tumors.1 The classification of Ewing

sarcoma was revolutionized with the identificationof a unique t(11;22) translocation, which is found inapproximately 85–90% of Ewing sarcoma cases,including those arising in both bone and softtissue.2–4 This translocation produces an oncogenicEWSR1-FLI1 fusion, combining the 5′ transcriptionalactivation domain of EWSR1 with the 3′ ETS DNA-binding domain of FLI1 to yield a potent oncogene.5

Several variant translocations have been implicatedin the minority of cases not exhibiting the typicalEWSR1-FLI1 fusion. These variants all pair the 5′portion of EWSR1 with the 3′ DNA-binding domainof a transcription factor (Figure 1). Of these rarervariants, the majority involve other ETS familytranscription factors, such as ERG (the most commonminor variant),6 ETV1,7 E1AF (ETV4),8 or FEV.9

More recently, even rarer fusion variants involvingtranscription factors without an ETS domain, such asNFATC2, have been discovered.10

Correspondence: Dr GW Charville, MD, PhD, Department ofPathology, Stanford University School of Medicine, 300 PasteurDrive, Lane 235, Stanford, CA 94305-5324, USA.E-mail: [email protected] 4 March 2017; revised 4 May 2017; accepted 21 May2017; published online 23 June 2017

Modern Pathology (2017), 1–9

© 2017 USCAP, Inc All rights reserved 0893-3952/17 $32.00 1

www.modernpathology.org

Page 2: EWSR1 fusion proteins mediate PAX7 expression in Ewing sarcomamed.stanford.edu/labs/vanderijn-west/Publications/modpathol201749a… · EWSR1 fusion proteins mediate PAX7 expression

Despite the identification of a pathognomonicmolecular lesion in Ewing sarcoma, immunohisto-chemical analysis of protein expression patternsremains an important adjunctive diagnostic tool,especially in resource-limited settings, cases with scantbiopsy material, or contexts in which there is clinicalurgency and the relative availability of immunohisto-chemistry enables more rapid diagnosis.11 Even beforethe identification of prototypical EWSR1 transloca-tions, it was noted that Ewing sarcoma expresses CD99(MIC2). Although it is highly sensitive for theidentification of Ewing sarcoma, CD99 is alsoexpressed in other neoplasms, including synovialsarcoma,12,13 mesenchymal chondrosarcoma,14 malig-nant peripheral nerve sheath tumor,13 plasma cellmyeloma,15 neuroendocrine tumors,16 and lympho-blastic lymphoma,17,18 among several others. In addi-tion, the degree of CD99 membranous accentuationcan vary from case-to-case, perhaps due to pre-analyticvariables such as time to fixation or time of fixation.Immunohistochemistry using an antibody recognizingthe ETS domain of FLI1 has also proven to besomewhat useful in the identification of Ewingsarcoma, particularly when combined with CD99analyses.19–21 Much like CD99, however, FLI1 expres-sion is not specific to Ewing sarcoma, showing exp-ression in normal endothelium and lymphocytes,19along with the vast majority of lymphoblastic lympho-mas and vascular neoplasms.21–23 FLI1 is alsoexpressed in smaller fractions of several other neo-plasms, such as squamous cell carcinoma, urothelialcarcinoma, Merkel cell carcinoma, lung adenocarci-noma, and melanoma.21,23 Akin to FLI1 immunohis-tochemical analysis, immunohistochemistry using an

antibody recognizing ERG is able to distinguish thesubset of Ewing sarcoma with an underlying EWSR1-ERG fusion.24

Recent advances in the genetic classification ofundifferentiated sarcomas has led to the identifica-tion of a molecularly distinct Ewing-like tumorbearing a recurrent translocation linking the CICgene with the DUX4 or paralogous DUX4L gene(Figure 1).25–27 In addition to having an indistinctround cell morphology, the CIC-DUX4 sarcomasshare many of the immunohistochemical features ofEwing sarcoma, including expression of CD99 andFLI1.28 Although these classical immunohistochem-ical markers do not distinguish closely related roundcell sarcomas, recent studies have identified severalnovel markers. NKX2-2 is a transcription factorexpressed in over 90% of Ewing sarcomas ando5% of CIC-DUX4 sarcomas; however, NKX2-2expression is not entirely specific for Ewing sarcoma,with significant expression in mesenchymal chon-drosarcoma and olfactory neuroblastoma, amongother tumors.29,30 Conversely, the transcriptionfactors WT1 and ETV4 are expressed in most CIC-DUX4 sarcomas, but not in Ewing sarcoma.31,32Moreover, just as FLI1 or ERG expression is notedin ES cases bearing translocations involving theseproteins, DUX4 expression has recently been shownto be a fusion-specific marker of CIC-DUX4sarcoma.33

In a previous study of PAX7 expression in an arrayof bone and soft tissue tumors, we observed that, inaddition to its expression in rhabdomyosarcoma,PAX7 was also expressed in a small cohort of sevenEwing sarcoma samples.34 Here, we aim to expandon this finding by exploring the diagnostic signifi-cance of PAX7 expression in a large series of Ewingsarcoma cases, along with several CIC-DUX4 sarco-mas, and by studying the mechanism of PAX7expression in Ewing sarcoma using curatedhigh-throughput sequencing data.

Materials and methods

Cases of Ewing sarcoma, numbering 103 in all, weregathered from the surgical pathology files of TheUniversity of Texas MD Anderson Cancer Center,University of Michigan Health System, and StanfordUniversity Medical Center. Seventy-three cases ofEwing sarcoma were evaluated as duplicate tissuecores represented as five micron-thick sections of atissue microarray. An additional 17 Ewing sarcomacases were evaluated as five micron-thick sectionsprepared from formalin-fixed paraffin-embeddedwhole-tissue sections. Thirteen Ewing sarcoma caseswere evaluated as single tissue cores represented asfive micron-thick sections of a separate tissuemicroarray. Twenty cases of CIC-DUX4 sarcomawere retrieved from the surgical pathology andconsultation files of Brigham and Women’s Hospitaland were evaluated as four micron-thick sections

Figure 1 Schematic of translocations involved in Ewing sarcomaand CIC-DUX sarcoma. Representative illustrations of the EWSR1-containing Ewing sarcoma translocations in this study highlightthe N-terminal transcriptional activation domain (TAD) of EWSR1.While translocations with FLI1 and ERG bear C-terminal E26transformation-specific (ETS) DNA-binding domains, the translo-cation with NFATc2 carries a DNA-binding Rel homology domain(RHD) at its C terminus. The CIC-DUX4 translocation includes ahigh mobility group DNA-binding domain at the N terminus ofCIC. Known functional domains of DUX4, including twoDNA-binding homeodomains, are not present in the translocation.

PAX7 expression in Ewing sarcoma

2 GW Charville et al

Modern Pathology (2017) 00, 1–9

Page 3: EWSR1 fusion proteins mediate PAX7 expression in Ewing sarcomamed.stanford.edu/labs/vanderijn-west/Publications/modpathol201749a… · EWSR1 fusion proteins mediate PAX7 expression

prepared from formalin-fixed paraffin-embeddedwhole-tissue sections. An additional seven CIC-DUX4 cases from University of Michigan HealthSystem, Cleveland Clinic, and Cedars-Sinai MedicalCenter were evaluated as duplicate tissue coresrepresented as five micron-thick sections on a tissuemicroarray.28 Classification of all CIC-DUX4 sarco-mas was previously performed using fluorescentin situ hybridization (FISH) and/or reversetranscription-PCR (RT-PCR).28,31

Immunohistochemical detection of PAX7 and ERGwas performed as previously described.24,34 Afterantigen retrieval, histologic sections were stainedwith antibodies according to standard protocols forformalin-fixed paraffin-embedded material. A mono-clonal antibody against PAX7 raised in mouse waspurchased from the Developmental Studies Hybri-doma Bank (Iowa City, IA, USA). Anti-PAX7 anti-body from hybridoma supernatant was used at adilution of 1:200 (by volume) following antigenretrieval using 0.01M citrate buffer at pH 6.0. Amonoclonal antibody against ERG (clone EP111)raised in rabbit was purchased from Epitomics(Burlingame, CA, USA) and used at a dilution of1:2000 (by volume). Immunohistochemical analysisof NKX2-2 was performed essentially as previouslydescribed.29 A monoclonal antibody against NKX2-2raised in mouse was purchased from the Develop-mental Studies Hybridoma Bank (clone 74.5A5) andused at a dilution of 1:25 (by volume) followingpressure cooker antigen retrieval using 0.01M citratebuffer at pH 6.0. For PAX7, NKX2-2, and ERG,immunoreactivity was considered positive if greaterthan 20% of tumor cells showed staining withappropriate nuclear chromogen localization. Inten-sity of PAX7 staining was approximated semi-quantitatively using a scale akin to that commonlyused for estimation of estrogen receptor expressionin breast carcinoma: 1+ (weak but detectable abovecontrol), 2+ (distinct), and 3+ (strong).35 All caseswere studied according the ethical rules of eachinstitution and processed in an anonymous codedfashion.

Publically curated microarray, chromatinimmunoprecipitation-sequencing (ChIP-seq), andRNA-sequencing (RNA-seq) data were accessed viathe Gene Expression Omnibus (GEO) betweenAugust 15, 2016, and November 20, 2016. Pairedend RNA-seq reads from SKNMC and A673 cell lineswith control GFP-targeting short hairpin-RNA(shRNA) or FLI1-targeting shRNA (GSE61950)36were aligned to the hg19 genome assembly andquantified as the number of sequenced fragments perkilobase of exon per million fragments mapped(FPKM) using the Tuxedo protocol.37 All ChIP-seqdata files (GSE61953) were aligned to the hg19genome assembly and visualized using the Integra-tive Genomics Viewer. The y-axis scale was set at aconstant value for all samples of a given ChIP-seqexperiment. Microarray data sets (GSE60740) wereanalyzed using the Partek Genomics Suite (version

6.6) following exclusion of outlying samples byprincipal component analysis.

To characterize the status of the EWSR1 locus,total RNA was extracted from 3 formalin-fixedparaffin-embedded (FFPE) PAX7-negative cases andquantified using the Qubit RNA assay (ThermoFisher Scientific, Waltham, MA). Next generationsequencing (NGS) libraries were prepared usinganchored multiplex PCR-based methodology as pre-viously described,38 using the Archer FusionPlexSarcoma kit (ArcherDX Inc, Boulder, CO, USA). TheArcher FusionPlex Sarcoma assay targets 127 exonsin 26 genes for agnostic detection of gene fusionswithout a priori knowledge of the partner gene. NGSlibraries were generated according to the manufac-turer’s protocol using 100 ng of total RNA. Amplifi-able cDNA quality was assessed using the PreSeqRNA QC assay; 2 of 3 cases had amplifiable RNAbased on the PreSeq assay. Final NGS libraries werequantified using the KAPA Library QuantificationKit for the Illumina platform. Illumina paired endindexed libraries were sequenced 8-plex on a MiSeq(2 × 150 bp, v2) and data analyzed on a vendor-provided virtual-machine based analysis pipelinewith custom-developed output scripts. On average,sequencing with the Archer FusionPlex SarcomaPanel generated a total of 1.07 million paired endreads with approximately 28 980 unique RNA readsper sample. Sequencing quality was assessed by %purity filtered (PF) reads, % bases 4Q30, totalnumber of reads, total number of unique RNA reads,and percent of aligned reads with high-mappingquality.

Results

To explore differential PAX7 expression in Ewingsarcoma and CIC-DUX4 sarcoma, we first performedanalyses of microarray-based genome-wide geneexpression data from 12 CIC-DUX4 and 6 Ewingsarcoma samples, the latter bearing the variantEWSR1-NFATC2 translocation (GEO data seriesGSE60740). Among 54675 analyzed genetic loci, weidentified 7090 genes showing differential expressionbetween Ewing sarcoma and CIC-DUX4 cases, using afold-change of greater than two and a false discoveryrate (FDR) of less than 0.01 to define significantlydifferential expression (Figure 2a). Ranking thedifferentially expressed genes according to p-valuein a comparison of CIC-DUX4 sarcoma versus Ewingsarcoma (Figure 2b), we identified eight transcriptionfactors among the 50 genes with the lowest p-values:FOXG1,NR5A2, SOX5, VDR,WT1, EBF2, TFAP2, andPAX7. Among these, FOXG1, NR5A2, SOX5, VDR,TFAP2, and PAX7 showed increased expression inEwing sarcoma, while WT1 and EBF2 showedincreased expression in CIC-DUX4 sarcoma. PAX7exhibited a 73-fold increased expression in Ewingsarcoma (P=4.0×10−15); WT1 exhibited a 511-foldincreased expression in CIC-DUX4 sarcoma

Modern Pathology (2017) 00, 1–9

PAX7 expression in Ewing sarcoma

GW Charville et al 3

Page 4: EWSR1 fusion proteins mediate PAX7 expression in Ewing sarcomamed.stanford.edu/labs/vanderijn-west/Publications/modpathol201749a… · EWSR1 fusion proteins mediate PAX7 expression

(P=1.7×10−16). Consistent with prior studies,29,30NKX2-2 expression was significantly increased inEwing sarcoma (98-fold, P=9.7×10−9), while ETV4expression was significantly increased in CIC-DUX4sarcoma (12-fold, P=1.3×10−8). Interestingly, CD99showed only two-fold increased expression in Ewingsarcoma, a difference that did not reach statisticalsignificance (P40.05). Given that PAX7 is requiredfor the specification of adult skeletal muscle progeni-tors during mouse developmental myogenesis andserves as a marker of human adult skeletal muscleprogenitors,39,40 we also examined the expression ofstandard markers of myogenic differentiation.MYOD1, MYOG, and DES each showed less than1.25-fold differential expression in Ewing sarcomaversus CIC-DUX4 sarcoma (P40.05). Taken together,our analysis of whole-genome gene expression sug-gests that PAX7 is a marker of Ewing sarcoma thatdistinguishes it from CIC-DUX4 sarcoma.

To further study the diagnostic utility of PAX7expression in Ewing sarcoma, we used immunohis-tochemistry to analyze PAX7 protein in 103 Ewingsarcoma specimens. All specimens represented truebiologic replicates from distinct patients and werepreviously diagnosed as Ewing sarcoma using acombination of morphologic, immunohistochemical,and molecular features. Of the 103 evaluated cases,102 (99.0%) were positive for PAX7 (Figure 3a). Thelone PAX7-negative Ewing sarcoma case had noobvious distinguishing clinical or morphologic char-acteristic. Using a scale of 1–3+ to classify intensityof chromogen staining,35 98 out of 103 PAX7-positive cases (95.1%) demonstrated 2–3+ stainingin greater than 50% of Ewing sarcoma cell nuclei,with the vast majority having anti-PAX7 immunor-eactivity in 90–100% of cells. In the remaining five

Figure 2 Whole-genome gene expression analyses identify PAX7 expression in Ewing sarcoma. (a) Heat-map of 7090 differentiallyexpressed (42-fold increased or decreased, FDR o0.01) genes in Ewing sarcoma versus CIC-DUX4 sarcoma. 12 CIC-DUX4 sarcomas and 6EWSR1-NFATc2 Ewing sarcomas were clustered according to microarray-based gene expression signatures (GSE60740). (b) Volcano plotshowing P-value versus fold-change in a comparison of high-throughput microarray gene expression data from 12 CIC-DUX4 sarcomasamples and 6 EWSR1-NFATc2 Ewing sarcoma samples. Each point on the plot represents one of 54675 individually analyzed loci. WT1and PAX7 are highlighted as genes with significantly enriched expression in CIC-DUX4 sarcoma and Ewing sarcoma, respectively.

Figure 3 Strong and diffuse PAX7 expression in Ewing sarcoma,including EWSR1-FLI1 and EWSR1-ERG variants. (a) Representa-tive immunohistochemical detection of PAX7 expression in Ewingsarcoma. (b) Representative immunohistochemical analysis ofERG and PAX7 expression in Ewing sarcoma. Shown are examplesof Ewing sarcoma with the EWSR1-ERG fusion (top), which showsdiffuse anti-ERG immunoreactivity, and the EWSR1-FLI1 fusion(bottom), which shows no anti-ERG immunoreactivity.

Modern Pathology (2017) 00, 1–9

PAX7 expression in Ewing sarcoma

4 GW Charville et al

Page 5: EWSR1 fusion proteins mediate PAX7 expression in Ewing sarcomamed.stanford.edu/labs/vanderijn-west/Publications/modpathol201749a… · EWSR1 fusion proteins mediate PAX7 expression

PAX7-positive cases, there was, at minimum, 1–2+staining in ~20–25% of cells. In all five cases forwhich a metastasis from the primary tumor wasavailable for parallel immunohistochemical analysis,concordant PAX7 expression was identified. We,therefore, conclude that immunohistochemicaldetection of anti-PAX7 immunoreactivity is a sensi-tive marker of Ewing sarcoma.

In the initial cohort of Ewing sarcoma cases, a totalof three PAX7-negative samples were identified. Tofurther explore the molecular basis of PAX7 expres-sion in Ewing sarcoma, and to confirm the diagnosisof Ewing sarcoma in these rare PAX7-negative cases,we performed multiplexed sequencing of PCR-enriched candidate pathogenic gene fusions, includ-ing those involving EWSR1 (see Materials andMethods). This approach yielded informativesequencing libraries in two of three cases. Of thetwo sequenced cases, neither demonstrated anEwing sarcoma-associated translocation of EWSR1,or any other likely pathogenic fusion. Given the lackof molecular support for the diagnosis of Ewingsarcoma, the two specimens evaluated by thissequencing approach were excluded from analysesof PAX7 expression and all subsequent analyses.Although an additional PAX7-negative specimenwas not excluded given a failed sequencing analysis,it is important to note that there is no molecularconfirmation of the diagnosis in this lone PAX7-negative case.

Noting that NKX2-2 is a transcription factor withrobust expression in Ewing sarcoma, we wereinterested in studying its expression characteristicsin parallel with those of PAX7. Consistent withprevious reports,29 NKX2-2 expression was positive(defined as expression in more than 20% of cells) in84.9% (62/73) of cases. Four cases showed restrictedexpression in 10–20% of cells. Interestingly, thesingle PAX7-negative Ewing sarcoma case alsolacked NKX2-2 expression. These findings, togetherwith our earlier observations, indicate that there isan immunohistochemically distinct PAX7-positive,NKX2-2-negative subtype of Ewing sarcoma.

As the EWSR1-FLI1 translocation accounts forapproximately 85–90% of Ewing sarcoma, we nextsought to characterize PAX7 expression in the rarer

EWSR1-ERG molecular subtype. We identifiedEWSR1-ERG variants among 73 Ewing sarcoma testcases using anti-ERG immunohistochemistry.24 Inall, we discovered five cases showing 2+ nuclearanti-ERG immunoreactivity in greater than 90% ofcells (5/73, 6.8%), indicating an underlying EWSR1-ERG translocation (Figure 3b). The remaining 68analyzed cases showed no significant ERG immu-nostaining within tumor cells, with vascularendothelial cells serving as a positive internalcontrol (Figure 3b). We found that PAX7 wasexpressed by immunohistochemistry in all 5 ERG-positive Ewing sarcoma cases (5/5, 100%). NKX2-2was also expressed in each of these cases. Thesefindings indicate that PAX7 expression is a marker ofboth common and variant forms of Ewing sarcoma.

Our analyses of genome-wide gene expression dataindicated not only that PAX7 expression is a markerof Ewing sarcoma, but also that it distinguishesEwing sarcoma from CIC-DUX4 sarcoma. We, there-fore, sought to evaluate PAX7 expression in a seriesof molecularly confirmed CIC-DUX4 sarcomas.Immunohistochemical analysis of PAX7 expressionin 27 cases of CIC-DUX4 sarcoma revealed nosignificant expression in any case (0/27; Figure 4).These immunohistochemical studies confirm thatPAX7 expression differentiates Ewing sarcoma fromCIC-DUX4 sarcoma.

These observations, in conjunction with ourprevious studies,34 suggest that significant PAX7expression is a unique feature of rhabdomyosarcomaand Ewing sarcoma. In contrast to rhabdomyosar-coma, Ewing sarcoma does not show myogenicdifferentiation, as manifested by a lack of expressionof myogenic genes, such as DES, in the large majorityof cases.11 What accounts for the observation ofrobust PAX7 expression in Ewing sarcoma? Toaddress this question, we turned to a publishedstudy that used high-throughput RNA-sequencing(RNA-seq), along with chromatin immunopre-cipitation-sequencing (ChIP-seq), to map EWSR1-FLI1-dependent chromatin remodeling and asso-ciated gene expression changes in Ewingsarcoma.36 In this prior study, regulation of thePAX7 gene was not specifically addressed and we,therefore, focused our analyses on PAX7 using thislarge sequencing data set. RNA-seq studies of twoEwing sarcoma cell lines (SKNMC and A673) in thisstudy showed expression of PAX7 in both lines(Figure 5a). Notably, expression of PAX7 in bothEwing sarcoma cell lines was dependent on expres-sion of the EWSR1-FLI1 fusion gene as evidenced bydecreased PAX7 transcript levels in the setting ofshRNA-mediated EWSR1-FLI1 knockdown(Figure 5a). The pattern of PAX7 expression inEwing sarcoma cells lines mimicked that of NKX2-2 (Figure 5a), which is known to be an EWSR1-FLI1-activated gene in Ewing sarcoma.36 Despite robustEWSR1-FLI1-dependent PAX7 expression in Ewingsarcoma cells, MYOD1, MYOG, and DES showed nosignificant expression by RNA-seq in Ewing sarcoma

Figure 4 Absence of PAX7 expression in CIC-DUX4 sarcoma.Representative H&E and immunohistochemical analysis of PAX7expression in a CIC-DUX4 sarcoma case.

Modern Pathology (2017) 00, 1–9

PAX7 expression in Ewing sarcoma

GW Charville et al 5

Page 6: EWSR1 fusion proteins mediate PAX7 expression in Ewing sarcomamed.stanford.edu/labs/vanderijn-west/Publications/modpathol201749a… · EWSR1 fusion proteins mediate PAX7 expression

(Figure 5a), in keeping with the lack of myogenicdifferentiation generally seen in this tumor.11 Riggiet al. have previously shown that EWSR1-FLI1 bindsat GGAA repeat elements in the genome to induce

chromatin opening at de novo transcriptionalenhancers.36 Given that RNA-seq data revealedEWSR1-FLI1-dependent PAX7 expression in Ewingsarcoma, we hypothesized that the fusion protein

Figure 5 Identification of a PAX7 enhancer activated by EWSR1-FLI1. (a) Quantification of expression of NKX2-2, PAX7, MYOD1,MYOG,and DES by RNA-sequencing in two control (shGFP) and EWSR1-FLI1-deficient (shFLI1) Ewing sarcoma cell lines (SKNMC and A673).Relative expression values are quantified as fragments per kilobase of transcript per million mapped reads (FPKM). (b) ChIP-sequencinganalysis in SKNMC Ewing sarcoma cells identifies an EWSR1-FLI1-bound enhancer corresponding to a site of H3K27 acetylation proximalto the PAX7 locus. This site corresponds to a consensus EWS-FLI1-binding site sequence with GGAA repeats. EWS-FLI1 binding andcorresponding H3K27 acetylation is lost upon knockdown of EWS-FLI1 expression (shFLI1). (c) ChIP-sequencing analysis in A673 Ewingsarcoma cells identifies an EWSR1-FLI1-bound enhancer corresponding to a site of H3K27 acetylation proximal to the PAX7 locus,analogous to that identified in SKNMC cells. Again, EWSR1-FLI1 binding and corresponding H3K27 acetylation is lost upon knockdownof EWSR1-FLI1 expression (shFLI1). (d) ChIP-sequencing demonstrates gain of H3K27 acetylation at the EWSR1-FLI1-bound PAX7enhancer in mesenchymal stem cells (MSCs) upon overexpression of an EWSR1-FLI1 transgene. (e) ChIP-sequencing analysis identifies apeak of H3K27 acetylation at the site of an EWSR1-FLI1-bound PAX7 enhancer in primary Ewing sarcoma. Three biologically independentEwing sarcoma samples are shown.

Modern Pathology (2017) 00, 1–9

PAX7 expression in Ewing sarcoma

6 GW Charville et al

Page 7: EWSR1 fusion proteins mediate PAX7 expression in Ewing sarcomamed.stanford.edu/labs/vanderijn-west/Publications/modpathol201749a… · EWSR1 fusion proteins mediate PAX7 expression

bound an activated enhancer of PAX7. AnalyzingEWSR1-FLI1 genome-wide ChIP-seq data, we identi-fied a discrete EWSR1-FLI1 binding site, locatedapproximately 20 kb upstream of the PAX7 promo-ter, which was present in both SKNMC and A673Ewing sarcoma cell lines in analyses of anti-FLI1ChIP-seq data (Figures 5b and c). This binding sitecoincided with a peak of regulatory H3K27 acetyla-tion (Figures 5b and c), providing support for itscharacterization as a transcriptional enhancer. Con-sistent with current models of EWSR1-FLI1 enhanceractivation, the position of EWSR1-FLI1 binding andH3K27 acetylation also coincided with a consensusbinding sequence of 18 GGAA nucleotide repeats(Figures 5b and c). In support of a model of PAX7enhancer activation by EWSR1-FLI1, shRNA-mediated suppression of EWSR1-FLI1 levels in bothSKNMC and A673 cells caused concordant loss ofH3K27 acetylation at the site of fusion proteinbinding (Figures 5b and c). These data indicate thatEWSR1-FLI1 is necessary for H3K27 acetylation of aPAX7 enhancer in Ewing sarcoma. As evidence thatEWSR1-FLI1 is also sufficient for activation of thePAX7 enhancer, ChIP-seq analyses of culturedmesenchymal stem cells (MSCs) showed gain ofregulatory H3K27 acetylation at the aforementionedEWSR1-FLI1 binding site in MSCs upon exogenousviral expression of EWSR1-FLI1 (Figure 5d). ChIP-seq analyses of genome-wide H3K27 acetylation inthree primary Ewing sarcoma samples showed thatthe same peak of H3K27 acetylation is present at thecandidate PAX7 enhancer in these primary tumorsamples (Figure 5e). Taken together, these analysesprovide evidence that a cis regulatory mechanism ofEWSR1-FLI1 binding and enhancer activation leadsto expression of PAX7 in Ewing sarcoma.

Discussion

The results of our studies provide support for theutility of immunohistochemical analysis of PAX7expression as a diagnostic marker of Ewing sarcoma.The overall sensitivity of PAX7 expression fordetection of Ewing sarcoma in 103 previouslydiagnosed cases was 99.0% in this study. Amongpositive cases, the vast majority showed 2–3+staining in greater than 50% of cells. In previouswork,34 we have shown that, among a diverse arrayof soft tissue and small round blue-cell neoplasms,expression of PAX7 is limited to rhabdomyosarcomaand Ewing sarcoma. These previous analysesincluded additional EWSR1 translocation-driventumors—extraskeletal myxoid chondrosarcoma anddesmoplastic small round cell tumor—both of whichshowed no PAX7 expression.

Support for the use of PAX7 immunohistochem-istry in the diagnosis of Ewing sarcoma is providedby our analyses of high-throughput sequencing datato arrive at a mechanism that accounts for EWSR1fusion protein-dependent PAX7 expression.

Assessment of whole-genome EWSR1-FLI1 ChIP-seq data shows binding of the fusion protein at aposition 5′ to the PAX7 promoter. This positioncoincides with a consensus GGAA repeat-containingEWSR1-FLI1 binding sequence and a peak ofacetylated H3K27, consistent with an activatedenhancer. These data support a model in whichtranscription of PAX7 is activated by binding of theoncogenic EWSR1-FLI1 fusion protein at a cisregulatory element. To further support this model,we find that knockdown of EWSR1-FLI1 expressionconcordantly abrogates both H3K27 acetylation atthe PAX7 enhancer and PAX7 expression in Ewingsarcoma cells.

In practice, the application of PAX7 immunohis-tochemistry can be used to narrow a fairly broaddifferential diagnosis of a morphologically undiffer-entiated small round cell or spindle-cell tumor. If aPAX7-expressing neoplasm is identified, distinguish-ing between Ewing sarcoma and rhabdomyosarcomamay be aided by additional immunohistochemicalmarkers. For instance, demonstration of myogenicdifferentiation by expression of DES, MYOD1, orMYOG would strongly favor rhabdomyosarcoma,though rare cases of Ewing sarcoma can exhibitmyogenic features (see Discussion below). Whileexpression of CD99 would typically favor Ewingsarcoma, CD99 expression has also been reported ina minor subset of embryonal rhabdomyosarcomas.41Similarly, expression of NKX2-2 would support thediagnosis of Ewing sarcoma.29 Still, identification ofEWSR1 translocation in ES or PAX3/7-FOXO1 trans-location in alveolar rhabdomyosarcoma will remainthe diagnostic ‘gold-standard’ for cases with ambig-uous immunohistochemical features. Even in themost challenging cases, we anticipate that immuno-histochemical markers, such as PAX7, will help tocurtail costs by enabling efficient triaging of cases forappropriate molecular studies.

PAX7 immunohistochemistry will also have utilityin a panel of antibodies for distinguishing Ewingsarcoma and CIC-DUX4 sarcoma. In our study, 99%of Ewing sarcoma expressed PAX7, whereas none of27 cases of CIC-DUX4 sarcoma showed evidence ofPAX7 expression. Although less is known about themolecular underpinnings of aberrant transcriptionalpathways in CIC-DUX4 sarcoma, this distinctionwith regard to PAX7 expression is presumably due toan absence of transcriptional activation of the PAX7locus by the CIC-DUX4 oncogenic fusion protein.Conversely, WT1 showed significantly increasedexpression in CIC-DUX4 sarcoma relative to Ewingsarcoma by whole-genome gene expression analysisperformed in this study, corroborating previousstudies and suggesting that paired analysis of PAX7and WT1 by immunohistochemistry can effectivelypredict the underlying genetic lesion in a largesubset of morphologically ambiguous round cellsarcomas. Specifically, WT1 immunohistochemicalanalyses have shown 100% specificity and at least95% sensitivity in distinguishing CIC-DUX4 tumors

Modern Pathology (2017) 00, 1–9

PAX7 expression in Ewing sarcoma

GW Charville et al 7

Page 8: EWSR1 fusion proteins mediate PAX7 expression in Ewing sarcomamed.stanford.edu/labs/vanderijn-west/Publications/modpathol201749a… · EWSR1 fusion proteins mediate PAX7 expression

from ES;31,32 thus, coupling analysis of WT1 withanalysis of PAX7, which is similarly sensitive andspecific, would make this particular distinction evenmore straightforward and robust.

In the absence of PAX7, muscle-resident adultstem cells more readily adopt a non-myogenic cellfate,40 suggesting a potential role for PAX7 intranscriptional regulation of the myogenic geneexpression program. Indeed, PAX7 binds regulatoryelements, induces local histone modifications, andactivates transcription of downstream myogenictarget genes, such as MYF5 and MYOD1.42,43Although ES cases only rarely exhibit significantmyogenic differentiation,11 several cases of so-called‘polyphenotypic’ or ‘hybrid’ tumors have beenreported.44,45 Given evidence that PAX7 has a rolein regulating myogenic differentiation, our observa-tion of EWSR1 translocation-dependent PAX7expression in Ewing sarcoma, offers an intriguingmolecular explanation for these polyphenotypictumors. At the same time, since PAX7 is expressedin the large majority of Ewing sarcoma cases, whataccounts for the relatively infrequent occurrence ofthe myogenic phenotype in Ewing sarcoma?Mechanisms that could explain this interestingfeature include expression of a repressor of myo-genic regulatory factor gene transcription or directinhibition of myogenic regulatory factor gene enhan-cers by the EWSR1 fusion protein itself. Futurestudies of myogenic regulatory factor expression inEwing sarcoma may provide additional insight.

Immunohistochemical studies have proven to beimportant tools in the diagnosis of soft tissuesarcomas defined by underlying genetic transloca-tions. To date, these immunohistochemical assayshave focused largely on the aberrant expression ofone of the genes involved in the pathogenictranslocation. Examples of such helpful immunohis-tochemical signatures include identification of WT1in desmoplastic small round cell tumor, STAT6 insolitary fibrous tumor, and, to some extent, FLI1 inEwing sarcoma. More recently, BCOR immunohis-tochemistry has been found to be a useful indicatorof underlying BCOR gene abnormalities.46 In con-trast to these approaches, the application of immu-nohistochemistry presented here for the diagnosis ofEwing sarcoma uniquely exploits a characteristictranscriptional readout of the disease-causing trans-location to aid in diagnosis. Presumably, uniquetranscriptional targets, or unique combinations oftargets, exist for other translocation-driven sarcomasand might be analyzed in a similar manner.Unbiased comparisons of whole-genome geneexpression data can help to identify these uniquesignatures, while mechanistic insights can be usedboth to justify the utility of a biomarker and accountfor features of a marker’s sensitivity and specificity.We anticipate that application of these approacheswill aid in discovery of a new class of biomarkers foran array of sarcomas.

Acknowledgments

The Amschwand Sarcoma Cancer Foundation isthanked for its generous support of this work(WLW, DRI, and AJL). This study was funded byDepartmental Research Funds.

Disclosure/conflict of interest

The authors declare no conflict of interest.

References

1 Carpentieri DF, Qualman SJ, Bowen J, et al. Protocol forthe examination of specimens from pediatric and adultpatients with osseous and extraosseous Ewing sarcomafamily of tumors, including peripheral primitive neu-roectodermal tumor and Ewing sarcoma. Arch PatholLab Med 2005;129:866–873.

2 de Chadarevian JP, Vekemans M, Seemayer TA.Reciprocal translocation in small-cell sarcomas. N EnglJ Med 1984;311:1702–1703.

3 Turc-Carel C, Aurias A, Mugneret F, et al. Chromo-somes in Ewing’s sarcoma. I. An evaluation of 85 casesof remarkable consistency of t(11;22)(q24;q12). CancerGenet Cytogenet 1988;32:229–238.

4 Whang-Peng J, Triche TJ, Knutsen T, et al. Chromo-some translocation in peripheral neuroepithelioma. NEngl J Med 1984;311:584–585.

5 Delattre O, Zucman J, Plougastel B, et al. Gene fusionwith an ETS DNA-binding domain caused by chromo-some translocation in human tumours. Nature 1992;359:162–165.

6 Sorensen PH, Lessnick SL, Lopez-Terrada D, et al. Asecond Ewing’s sarcoma translocation, t(21;22), fusesthe EWS gene to another ETS-family transcriptionfactor, ERG. Nat Genet 1994;6:146–151.

7 Jeon IS, Davis JN, Braun BS, et al. A variant Ewing’ssarcoma translocation (7;22) fuses the EWS gene to theETS gene ETV1. Oncogene 1995;10:1229–1234.

8 Urano F, Umezawa A, Yabe H, et al. Molecular analysisof Ewing’s sarcoma: another fusion gene, EWS-E1AF,available for diagnosis. Jpn J Cancer Res 1998;89:703–711.

9 Peter M, Couturier J, Pacquement H, et al. A newmember of the ETS family fused to EWS inEwing tumors. Oncogene 1997;14:1159–1164.

10 Szuhai K, Ijszenga M, de Jong D, et al. The NFATc2 geneis involved in a novel cloned translocation in a Ewingsarcoma variant that couples its function in immunologyto oncology. Clin Cancer Res 2009;15:2259–2268.

11 Folpe AL, Goldblum JR, Rubin BP, et al. Morphologicand immunophenotypic diversity in Ewing familytumors: a study of 66 genetically confirmed cases. AmJ Surg Pathol 2005;29:1025–1033.

12 Folpe AL, Schmidt RA, Chapman D, et al. Poorlydifferentiated synovial sarcoma: immunohistochemicaldistinction from primitive neuroectodermal tumors andhigh-grade malignant peripheral nerve sheath tumors.Am J Surg Pathol 1998;22:673–682.

13 Olsen SH, Thomas DG, Lucas DR. Cluster analysis ofimmunohistochemical profiles in synovial sarcoma,malignant peripheral nerve sheath tumor, and Ewingsarcoma. Mod Pathol 2006;19:659–668.

Modern Pathology (2017) 00, 1–9

PAX7 expression in Ewing sarcoma

8 GW Charville et al

Page 9: EWSR1 fusion proteins mediate PAX7 expression in Ewing sarcomamed.stanford.edu/labs/vanderijn-west/Publications/modpathol201749a… · EWSR1 fusion proteins mediate PAX7 expression

14 Granter SR, Renshaw AA, Fletcher CD, et al. CD99reactivity in mesenchymal chondrosarcoma. HumPathol 1996;27:1273–1276.

15 Shin S-J, Lee H, Jung G, et al. Expression of CD99 inmultiple myeloma: a clinicopathologic and immuno-histochemical study of 170 cases. Korean J Pathol2014;48:209–216.

16 Pelosi G, Fraggetta F, Sonzogni A, et al. CD99 immunor-eactivity in gastrointestinal and pulmonary neuroendo-crine tumours. Virchows Arch 2000;437:270–274.

17 Lucas DR, Bentley G, Dan ME, et al. Ewing sarcoma vslymphoblastic lymphoma. A comparative immuno-histochemical study. Am J Clin Pathol 2001;115:11–17.

18 Ozdemirli M, Fanburg-Smith JC, Hartmann DP, et al.Differentiating lymphoblastic lymphoma and Ewing’ssarcoma: lymphocyte markers and gene rearrangement.Mod Pathol 2001;14:1175–1182.

19 Folpe AL, Hill CE, Parham DM, et al. Immunohistochem-ical detection of FLI-1 protein expression: a study of 132round cell tumors with emphasis on CD99-positive mimicsof Ewing’s sarcoma/primitive neuroectodermal tumor. Am JSurg Pathol 2000;24:1657–1662.

20 Llombart-Bosch A, Navarro S. Immunohistochemicaldetection of EWS and FLI-1 proteinss in Ewing sarcomaand primitive neuroectodermal tumors: comparativeanalysis with CD99 (MIC-2) expression. Appl Immu-nohistochem Mol Morphol 2001;9:255–260.

21 Rossi S, Orvieto E, Furlanetto A, et al. Utility of theimmunohistochemical detection of FLI-1 expression inround cell and vascular neoplasm using a monoclonalantibody. Mod Pathol 2004;17:547–552.

22 Folpe AL, Chand EM, Goldblum JR, et al. Expression ofFli-1, a nuclear transcription factor, distinguishesvascular neoplasms from potential mimics. Am J SurgPathol 2001;25:1061–1066.

23 Mhawech-Fauceglia P, Herrmann FR, Bshara W, et al.Friend leukaemia integration-1 expression in malig-nant and benign tumours: a multiple tumour tissuemicroarray analysis using polyclonal antibody. J ClinPathol 2007;60:694–700.

24 Wang W-L, Patel NR, Caragea M, et al. Expression of ERG,an Ets family transcription factor, identifies ERG-rearrangedEwing sarcoma. Mod Pathol 2012;25:1378–1383.

25 Choi E-YK, Thomas DG, McHugh JB, et al. Undiffer-entiated small round cell sarcoma with t(4;19)(q35;q13.1) CIC-DUX4 fusion: a novel highly aggressive softtissue tumor with distinctive histopathology. Am J SurgPathol 2013;37:1379–1386.

26 Graham C, Chilton-MacNeill S, Zielenska M, et al. TheCIC-DUX4 fusion transcript is present in a subgroup ofpediatric primitive round cell sarcomas. Hum Pathol2012;43:180–189.

27 Italiano A, Sung YS, Zhang L, et al. High prevalence ofCIC fusion with double-homeobox (DUX4) transcrip-tion factors in EWSR1-negative undifferentiated smallblue round cell sarcomas. Genes Chromosomes Cancer2012;51:207–218.

28 Smith SC, Buehler D, Choi E-YK, et al. CIC-DUXsarcomas demonstrate frequent MYC amplification andETS-family transcription factor expression. Mod Pathol2015;28:57–68.

29 Hung YP, Fletcher CDM, Hornick JL. Evaluation ofNKX2-2 expression in round cell sarcomas and othertumors with EWSR1 rearrangement: imperfect specifi-city for Ewing sarcoma. Mod Pathol 2016;29:370–380.

30 Yoshida A, Sekine S, Tsuta K, et al. NKX2.2 is a usefulimmunohistochemical marker for Ewing sarcoma. Am JSurg Pathol 2012;36:993–999.

31 Hung YP, Fletcher CD, Hornick JL. Evaluation of ETV4and WT1 expression in CIC-rearranged sarcomas andhistologic mimics. Mod Pathol 2016;29:1324–1334.

32 Specht K, Sung Y-S, Zhang L, et al. Distinct transcrip-tional signature and immunoprofile of CIC-DUX4fusion-positive round cell tumors compared toEWSR1-rearranged Ewing sarcomas: further evidencetoward distinct pathologic entities. Genes Chromo-somes Cancer 2014;53:622–633.

33 Siegele B, Roberts J, Black JO, et al. DUX4 immuno-histochemistry is a highly sensitive and specific markerfor CIC-DUX4 fusion-positive round cell tumor. Am JSurg Pathol 2017;41:423–429.

34 Charville GW, Varma S, Forgó E, et al. PAX7 expres-sion in rhabdomyosarcoma, related soft tissue tumors,and small round blue cell neoplasms. Am J Surg Pathol2016;40:1305–1315.

35 McCarty KS, Szabo E, Flowers JL, et al. Use of amonoclonal anti-estrogen receptor antibody in theimmunohistochemical evaluation of human tumors.Cancer Res 1986;46:4244s–4248s.

36 Riggi N, Knoechel B, Gillespie SM, et al. EWS-FLI1utilizes divergent chromatin remodeling mechanismsto directly activate or repress enhancer elements inEwing sarcoma. Cancer Cell 2014;26:668–681.

37 Trapnell C, Roberts A, Goff L, et al. Differential geneand transcript expression analysis of RNA-seq experi-ments with TopHat and Cufflinks. Nat Protoc 2012;7:562–578.

38 Zheng Z, Liebers M, Zhelyazkova B, et al. Anchoredmultiplex PCR for targeted next-generation sequencing.Nat Med 2014;20:1479–1484.

39 Charville GW, Cheung TH, Yoo B, et al. Ex vivoexpansion and in vivo self-renewal of human musclestem cells. Stem Cell Rep 2015;5:621–632.

40 Seale P, Sabourin LA, Girgis-Gabardo A, et al. Pax7 isrequired for the specification of myogenic satellitecells. Cell 2000;102:777–786.

41 Ramani P, Rampling D, Link M. Immunocytochemicalstudy of 12E7 in small round-cell tumours of child-hood: an assessment of its sensitivity and specificity.Histopathology 1993;23:557–561.

42 Relaix F, Montarras D, Zaffran S, et al. Pax3 and Pax7have distinct and overlapping functions in adultmuscle progenitor cells. J Cell Biol 2006;172:91–102.

43 McKinnell IW, Ishibashi J, Le Grand F, et al.Pax7 activates myogenic genes by recruitment of ahistone methyltransferase complex. Nat Cell Biol 2008;10:77–84.

44 Barisella M, Collini P, Orsenigo M, et al.Unusual myogenic and melanocytic differentiation ofsoft tissue pPNETs: an immunohistochemical andmolecular study of 3 cases. Am J Surg Pathol 2010;34:1002–1006.

45 Sorensen PH, Shimada H, Liu XF, et al. Biphenotypicsarcomas with myogenic and neural differentiationexpress the Ewing’s sarcoma EWS/FLI1 fusion gene.Cancer Res 1995;55:1385–1392.

46 Kao Y-C, Sung Y-S, Zhang L, et al. BCOR over-expression is a highly sensitive marker in round cellsarcomas with BCOR genetic abnormalities. Am J SurgPathol 2016;40:1670–1678.

Modern Pathology (2017) 00, 1–9

PAX7 expression in Ewing sarcoma

GW Charville et al 9